neighbour.c

linux 内核源代码

			atomic_inc(&hh->hh_refcnt);
			hh->hh_next = n->hh;
			n->hh	    = hh;
			if (n->nud_state & NUD_CONNECTED)
				hh->hh_output = n->ops->hh_output;
			else
				hh->hh_output = n->ops->output;
		}
	}
	if (hh)	{
		atomic_inc(&hh->hh_refcnt);
		dst->hh = hh;
	}
}

/* This function can be used in contexts, where only old dev_queue_xmit
   worked, f.e. if you want to override normal output path (eql, shaper),
   but resolution is not made yet.
 */

int neigh_compat_output(struct sk_buff *skb)
{
	struct net_device *dev = skb->dev;

	__skb_pull(skb, skb_network_offset(skb));

	if (dev_hard_header(skb, dev, ntohs(skb->protocol), NULL, NULL,
			    skb->len) < 0 &&
	    dev->header_ops->rebuild(skb))
		return 0;

	return dev_queue_xmit(skb);
}

/* Slow and careful. */

int neigh_resolve_output(struct sk_buff *skb)
{
	struct dst_entry *dst = skb->dst;
	struct neighbour *neigh;
	int rc = 0;

	if (!dst || !(neigh = dst->neighbour))
		goto discard;

	__skb_pull(skb, skb_network_offset(skb));

	if (!neigh_event_send(neigh, skb)) {
		int err;
		struct net_device *dev = neigh->dev;
		if (dev->header_ops->cache && !dst->hh) {
			write_lock_bh(&neigh->lock);
			if (!dst->hh)
				neigh_hh_init(neigh, dst, dst->ops->protocol);
			err = dev_hard_header(skb, dev, ntohs(skb->protocol),
					      neigh->ha, NULL, skb->len);
			write_unlock_bh(&neigh->lock);
		} else {
			read_lock_bh(&neigh->lock);
			err = dev_hard_header(skb, dev, ntohs(skb->protocol),
					      neigh->ha, NULL, skb->len);
			read_unlock_bh(&neigh->lock);
		}
		if (err >= 0)
			rc = neigh->ops->queue_xmit(skb);
		else
			goto out_kfree_skb;
	}
out:
	return rc;
discard:
	NEIGH_PRINTK1("neigh_resolve_output: dst=%p neigh=%p\n",
		      dst, dst ? dst->neighbour : NULL);
out_kfree_skb:
	rc = -EINVAL;
	kfree_skb(skb);
	goto out;
}

/* As fast as possible without hh cache */

int neigh_connected_output(struct sk_buff *skb)
{
	int err;
	struct dst_entry *dst = skb->dst;
	struct neighbour *neigh = dst->neighbour;
	struct net_device *dev = neigh->dev;

	__skb_pull(skb, skb_network_offset(skb));

	read_lock_bh(&neigh->lock);
	err = dev_hard_header(skb, dev, ntohs(skb->protocol),
			      neigh->ha, NULL, skb->len);
	read_unlock_bh(&neigh->lock);
	if (err >= 0)
		err = neigh->ops->queue_xmit(skb);
	else {
		err = -EINVAL;
		kfree_skb(skb);
	}
	return err;
}

static void neigh_proxy_process(unsigned long arg)
{
	struct neigh_table *tbl = (struct neigh_table *)arg;
	long sched_next = 0;
	unsigned long now = jiffies;
	struct sk_buff *skb;

	spin_lock(&tbl->proxy_queue.lock);

	skb = tbl->proxy_queue.next;

	while (skb != (struct sk_buff *)&tbl->proxy_queue) {
		struct sk_buff *back = skb;
		long tdif = NEIGH_CB(back)->sched_next - now;

		skb = skb->next;
		if (tdif <= 0) {
			struct net_device *dev = back->dev;
			__skb_unlink(back, &tbl->proxy_queue);
			if (tbl->proxy_redo && netif_running(dev))
				tbl->proxy_redo(back);
			else
				kfree_skb(back);

			dev_put(dev);
		} else if (!sched_next || tdif < sched_next)
			sched_next = tdif;
	}
	del_timer(&tbl->proxy_timer);
	if (sched_next)
		mod_timer(&tbl->proxy_timer, jiffies + sched_next);
	spin_unlock(&tbl->proxy_queue.lock);
}

void pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p,
		    struct sk_buff *skb)
{
	unsigned long now = jiffies;
	unsigned long sched_next = now + (net_random() % p->proxy_delay);

	if (tbl->proxy_queue.qlen > p->proxy_qlen) {
		kfree_skb(skb);
		return;
	}

	NEIGH_CB(skb)->sched_next = sched_next;
	NEIGH_CB(skb)->flags |= LOCALLY_ENQUEUED;

	spin_lock(&tbl->proxy_queue.lock);
	if (del_timer(&tbl->proxy_timer)) {
		if (time_before(tbl->proxy_timer.expires, sched_next))
			sched_next = tbl->proxy_timer.expires;
	}
	dst_release(skb->dst);
	skb->dst = NULL;
	dev_hold(skb->dev);
	__skb_queue_tail(&tbl->proxy_queue, skb);
	mod_timer(&tbl->proxy_timer, sched_next);
	spin_unlock(&tbl->proxy_queue.lock);
}


struct neigh_parms *neigh_parms_alloc(struct net_device *dev,
				      struct neigh_table *tbl)
{
	struct neigh_parms *p = kmemdup(&tbl->parms, sizeof(*p), GFP_KERNEL);

	if (p) {
		p->tbl		  = tbl;
		atomic_set(&p->refcnt, 1);
		INIT_RCU_HEAD(&p->rcu_head);
		p->reachable_time =
				neigh_rand_reach_time(p->base_reachable_time);
		if (dev) {
			if (dev->neigh_setup && dev->neigh_setup(dev, p)) {
				kfree(p);
				return NULL;
			}

			dev_hold(dev);
			p->dev = dev;
		}
		p->sysctl_table = NULL;
		write_lock_bh(&tbl->lock);
		p->next		= tbl->parms.next;
		tbl->parms.next = p;
		write_unlock_bh(&tbl->lock);
	}
	return p;
}

static void neigh_rcu_free_parms(struct rcu_head *head)
{
	struct neigh_parms *parms =
		container_of(head, struct neigh_parms, rcu_head);

	neigh_parms_put(parms);
}

void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms)
{
	struct neigh_parms **p;

	if (!parms || parms == &tbl->parms)
		return;
	write_lock_bh(&tbl->lock);
	for (p = &tbl->parms.next; *p; p = &(*p)->next) {
		if (*p == parms) {
			*p = parms->next;
			parms->dead = 1;
			write_unlock_bh(&tbl->lock);
			if (parms->dev)
				dev_put(parms->dev);
			call_rcu(&parms->rcu_head, neigh_rcu_free_parms);
			return;
		}
	}
	write_unlock_bh(&tbl->lock);
	NEIGH_PRINTK1("neigh_parms_release: not found\n");
}

void neigh_parms_destroy(struct neigh_parms *parms)
{
	kfree(parms);
}

static struct lock_class_key neigh_table_proxy_queue_class;

void neigh_table_init_no_netlink(struct neigh_table *tbl)
{
	unsigned long now = jiffies;
	unsigned long phsize;

	atomic_set(&tbl->parms.refcnt, 1);
	INIT_RCU_HEAD(&tbl->parms.rcu_head);
	tbl->parms.reachable_time =
			  neigh_rand_reach_time(tbl->parms.base_reachable_time);

	if (!tbl->kmem_cachep)
		tbl->kmem_cachep =
			kmem_cache_create(tbl->id, tbl->entry_size, 0,
					  SLAB_HWCACHE_ALIGN|SLAB_PANIC,
					  NULL);
	tbl->stats = alloc_percpu(struct neigh_statistics);
	if (!tbl->stats)
		panic("cannot create neighbour cache statistics");

#ifdef CONFIG_PROC_FS
	tbl->pde = create_proc_entry(tbl->id, 0, init_net.proc_net_stat);
	if (!tbl->pde)
		panic("cannot create neighbour proc dir entry");
	tbl->pde->proc_fops = &neigh_stat_seq_fops;
	tbl->pde->data = tbl;
#endif

	tbl->hash_mask = 1;
	tbl->hash_buckets = neigh_hash_alloc(tbl->hash_mask + 1);

	phsize = (PNEIGH_HASHMASK + 1) * sizeof(struct pneigh_entry *);
	tbl->phash_buckets = kzalloc(phsize, GFP_KERNEL);

	if (!tbl->hash_buckets || !tbl->phash_buckets)
		panic("cannot allocate neighbour cache hashes");

	get_random_bytes(&tbl->hash_rnd, sizeof(tbl->hash_rnd));

	rwlock_init(&tbl->lock);
	init_timer(&tbl->gc_timer);
	tbl->gc_timer.data     = (unsigned long)tbl;
	tbl->gc_timer.function = neigh_periodic_timer;
	tbl->gc_timer.expires  = now + 1;
	add_timer(&tbl->gc_timer);

	init_timer(&tbl->proxy_timer);
	tbl->proxy_timer.data	  = (unsigned long)tbl;
	tbl->proxy_timer.function = neigh_proxy_process;
	skb_queue_head_init_class(&tbl->proxy_queue,
			&neigh_table_proxy_queue_class);

	tbl->last_flush = now;
	tbl->last_rand	= now + tbl->parms.reachable_time * 20;
}

void neigh_table_init(struct neigh_table *tbl)
{
	struct neigh_table *tmp;

	neigh_table_init_no_netlink(tbl);
	write_lock(&neigh_tbl_lock);
	for (tmp = neigh_tables; tmp; tmp = tmp->next) {
		if (tmp->family == tbl->family)
			break;
	}
	tbl->next	= neigh_tables;
	neigh_tables	= tbl;
	write_unlock(&neigh_tbl_lock);

	if (unlikely(tmp)) {
		printk(KERN_ERR "NEIGH: Registering multiple tables for "
		       "family %d\n", tbl->family);
		dump_stack();
	}
}

int neigh_table_clear(struct neigh_table *tbl)
{
	struct neigh_table **tp;

	/* It is not clean... Fix it to unload IPv6 module safely */
	del_timer_sync(&tbl->gc_timer);
	del_timer_sync(&tbl->proxy_timer);
	pneigh_queue_purge(&tbl->proxy_queue);
	neigh_ifdown(tbl, NULL);
	if (atomic_read(&tbl->entries))
		printk(KERN_CRIT "neighbour leakage\n");
	write_lock(&neigh_tbl_lock);
	for (tp = &neigh_tables; *tp; tp = &(*tp)->next) {
		if (*tp == tbl) {
			*tp = tbl->next;
			break;
		}
	}
	write_unlock(&neigh_tbl_lock);

	neigh_hash_free(tbl->hash_buckets, tbl->hash_mask + 1);
	tbl->hash_buckets = NULL;

	kfree(tbl->phash_buckets);
	tbl->phash_buckets = NULL;

	remove_proc_entry(tbl->id, init_net.proc_net_stat);

	free_percpu(tbl->stats);
	tbl->stats = NULL;

	kmem_cache_destroy(tbl->kmem_cachep);
	tbl->kmem_cachep = NULL;

	return 0;
}

static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
	struct net *net = skb->sk->sk_net;
	struct ndmsg *ndm;
	struct nlattr *dst_attr;
	struct neigh_table *tbl;
	struct net_device *dev = NULL;
	int err = -EINVAL;

	if (nlmsg_len(nlh) < sizeof(*ndm))
		goto out;

	dst_attr = nlmsg_find_attr(nlh, sizeof(*ndm), NDA_DST);
	if (dst_attr == NULL)
		goto out;

	ndm = nlmsg_data(nlh);
	if (ndm->ndm_ifindex) {
		dev = dev_get_by_index(net, ndm->ndm_ifindex);
		if (dev == NULL) {
			err = -ENODEV;
			goto out;
		}
	}

	read_lock(&neigh_tbl_lock);
	for (tbl = neigh_tables; tbl; tbl = tbl->next) {
		struct neighbour *neigh;

		if (tbl->family != ndm->ndm_family)
			continue;
		read_unlock(&neigh_tbl_lock);

		if (nla_len(dst_attr) < tbl->key_len)
			goto out_dev_put;

		if (ndm->ndm_flags & NTF_PROXY) {
			err = pneigh_delete(tbl, nla_data(dst_attr), dev);
			goto out_dev_put;
		}

		if (dev == NULL)
			goto out_dev_put;

		neigh = neigh_lookup(tbl, nla_data(dst_attr), dev);
		if (neigh == NULL) {
			err = -ENOENT;
			goto out_dev_put;
		}

		err = neigh_update(neigh, NULL, NUD_FAILED,
				   NEIGH_UPDATE_F_OVERRIDE |
				   NEIGH_UPDATE_F_ADMIN);
		neigh_release(neigh);
		goto out_dev_put;
	}
	read_unlock(&neigh_tbl_lock);
	err = -EAFNOSUPPORT;

out_dev_put:
	if (dev)
		dev_put(dev);
out:
	return err;
}

static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
	struct net *net = skb->sk->sk_net;
	struct ndmsg *ndm;
	struct nlattr *tb[NDA_MAX+1];
	struct neigh_table *tbl;
	struct net_device *dev = NULL;
	int err;

	err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL);
	if (err < 0)
		goto out;

	err = -EINVAL;
	if (tb[NDA_DST] == NULL)
		goto out;

	ndm = nlmsg_data(nlh);
	if (ndm->ndm_ifindex) {
		dev = dev_get_by_index(net, ndm->ndm_ifindex);
		if (dev == NULL) {
			err = -ENODEV;
			goto out;
		}

		if (tb[NDA_LLADDR] && nla_len(tb[NDA_LLADDR]) < dev->addr_len)
			goto out_dev_put;
	}

	read_lock(&neigh_tbl_lock);
	for (tbl = neigh_tables; tbl; tbl = tbl->next) {
		int flags = NEIGH_UPDATE_F_ADMIN | NEIGH_UPDATE_F_OVERRIDE;
		struct neighbour *neigh;
		void *dst, *lladdr;

		if (tbl->family != ndm->ndm_family)
			continue;
		read_unlock(&neigh_tbl_lock);

		if (nla_len(tb[NDA_DST]) < tbl->key_len)
			goto out_dev_put;
		dst = nla_data(tb[NDA_DST]);
		lladdr = tb[NDA_LLADDR] ? nla_data(tb[NDA_LLADDR]) : NULL;

		if (ndm->ndm_flags & NTF_PROXY) {
			struct pneigh_entry *pn;

			err = -ENOBUFS;
			pn = pneigh_lookup(tbl, dst, dev, 1);
			if (pn) {
				pn->flags = ndm->ndm_flags;
				err = 0;
			}
			goto out_dev_put;
		}

		if (dev == NULL)
			goto out_dev_put;

		neigh = neigh_lookup(tbl, dst, dev);
		if (neigh == NULL) {
			if (!(nlh->nlmsg_flags & NLM_F_CREATE)) {
				err = -ENOENT;
				goto out_dev_put;
			}

			neigh = __neigh_lookup_errno(tbl, dst, dev);
			if (IS_ERR(neigh)) {
				err = PTR_ERR(neigh);
				goto out_dev_put;
			}
		} else {
			if (nlh->nlmsg_flags & NLM_F_EXCL) {
				err = -EEXIST;
				neigh_release(neigh);
				goto out_dev_put;
			}

			if (!(nlh->nlmsg_flags & NLM_F_REPLACE))
				flags &= ~NEIGH_UPDATE_F_OVERRIDE;
		}

		err = neigh_update(neigh, lladdr, ndm->ndm_state, flags);
		neigh_release(neigh);
		goto out_dev_put;
	}

	read_unlock(&neigh_tbl_lock);
	err = -EAFNOSUPPORT;

out_dev_put:
	if (dev)
		dev_put(dev);
out:
	return err;
}

static int neightbl_fill_parms(struct sk_buff *skb, struct neigh_parms *parms)
{
	struct nlattr *nest;

	nest = nla_nest_start(skb, NDTA_PARMS);
	if (nest == NULL)
		return -ENOBUFS;

	if (parms->dev)
		NLA_PUT_U32(skb, NDTPA_IFINDEX, parms->dev->ifindex);

	NLA_PUT_U32(skb, NDTPA_REFCNT, atomic_read(&parms->refcnt));
	NLA_PUT_U32(skb, NDTPA_QUEUE_LEN, parms->queue_len);
	NLA_PUT_U32(skb, NDTPA_PROXY_QLEN, parms->proxy_qlen);
	NLA_PUT_U32(skb, NDTPA_APP_PROBES, parms->app_probes);
	NLA_PUT_U32(skb, NDTPA_UCAST_PROBES, parms->ucast_probes);
	NLA_PUT_U32(skb, NDTPA_MCAST_PROBES, parms->mcast_probes);
	NLA_PUT_MSECS(skb, NDTPA_REACHABLE_TIME, parms->reachable_time);
	NLA_PUT_MSECS(skb, NDTPA_BASE_REACHABLE_TIME,
		      parms->base_reachable_time);
	NLA_PUT_MSECS(skb, NDTPA_GC_STALETIME, parms->gc_staletime);
	NLA_PUT_MSECS(skb, NDTPA_DELAY_PROBE_TIME, parms->delay_probe_time);
	NLA_PUT_MSECS(skb, NDTPA_RETRANS_TIME, parms->retrans_time);
	NLA_PUT_MSECS(skb, NDTPA_ANYCAST_DELAY, parms->anycast_delay);
	NLA_PUT_MSECS(skb, NDTPA_PROXY_DELAY, parms->proxy_delay);
	NLA_PUT_MSECS(skb, NDTPA_LOCKTIME, parms->locktime);

	return nla_nest_end(skb, nest);

nla_put_failure:
	return nla_nest_cancel(skb, nest);
}

static int neightbl_fill_info(struct sk_buff *skb, struct neigh_table *tbl,
			      u32 pid, u32 seq, int type, int flags)
{
	struct nlmsghdr *nlh;
	struct ndtmsg *ndtmsg;

	nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ndtmsg), flags);
	if (nlh == NULL)
		return -EMSGSIZE;

	ndtmsg = nlmsg_data(nlh);